implemented model-based reflect agent program 2D

aima/agents.py

@@ -191,8 +191,8 @@ def rule_match(state, rules):
 # ______________________________________________________________________________
 
 
-loc_A, loc_B = (0, 0), (1, 0)  # The two locations for the Vacuum world
-
+loc_A, loc_B, loc_C, loc_D = (0, 0), (1, 0), (0, 1), (1, 1)  # The four locations for the Vacuum world
+locations = [loc_A, loc_B, loc_C, loc_D]
 
 def RandomVacuumAgent():
     """Randomly choose one of the actions from the vacuum environment.
@@ -203,7 +203,7 @@ def RandomVacuumAgent():
     >>> environment.status == {(1,0):'Clean' , (0,0) : 'Clean'}
     True
     """
-    return Agent(RandomAgentProgram(['Right', 'Left', 'Suck', 'NoOp']))
+    return Agent(RandomAgentProgram(['Right', 'Left','Up','Down', 'Suck', 'NoOp']))
 
 
 def TableDrivenVacuumAgent():
@@ -261,20 +261,25 @@ def ModelBasedVacuumAgent():
     >>> environment.status == {(1,0):'Clean' , (0,0) : 'Clean'}
     True
     """
-    model = {loc_A: None, loc_B: None}
+    model = {loc_A: None, loc_B: None, loc_C: None, loc_D: None}
 
     def program(percept):
         """Same as ReflexVacuumAgent, except if everything is clean, do NoOp."""
-        location, status = percept
-        model[location] = status  # Update the model here
-        if model[loc_A] == model[loc_B] == 'Clean':
+        # location, status = percept
+        # model[location] = status  # Update the model here
+        state = update_state(state, action, percept, model)
+        if model[loc_A] == model[loc_B] == model[loc_C] == model[loc_D] == 'Clean':
             return 'NoOp'
         elif status == 'Dirty':
             return 'Suck'
         elif location == loc_A:
             return 'Right'
         elif location == loc_B:
             return 'Left'
+        elif location == loc_C:
+            return 'Down'
+        elif location == loc_D:
+            return 'Up'
 
     return Agent(program)
 
@@ -806,8 +811,7 @@ class TrivialVacuumEnvironment(Environment):
 
     def __init__(self):
         super().__init__()
-        self.status = {loc_A: random.choice(['Clean', 'Dirty']),
-                       loc_B: random.choice(['Clean', 'Dirty'])}
+        self.status = {loc: random.choice(['Clean', 'Dirty']) for loc in locations}
 
     def thing_classes(self):
         """Return the Thing/Agent classes that may populate this vacuum world."""
@@ -820,11 +824,18 @@ def percept(self, agent):
     def execute_action(self, agent, action):
         """Change agent's location and/or location's status; track performance.
         Score 10 for each dirt cleaned; -1 for each move."""
+        a, b = agent.location
         if action == 'Right':
-            agent.location = loc_B
+            agent.location = (a + 1, b)
             agent.performance -= 1
         elif action == 'Left':
-            agent.location = loc_A
+            agent.location = (a - 1, b)
+            agent.performance -= 1
+        elif action == 'Up':
+            agent.location = (a, b + 1)
+            agent.performance -= 1
+        elif action == 'Down':
+            agent.location = (a, b - 1)
             agent.performance -= 1
         elif action == 'Suck':
             if self.status[agent.location] == 'Dirty':
@@ -833,7 +844,7 @@ def execute_action(self, agent, action):
 
     def default_location(self, thing):
         """Agents start in either location at random."""
-        return random.choice([loc_A, loc_B])
+        return random.choice([loc_A, loc_B, loc_C, loc_D])
 
 
 # ______________________________________________________________________________

aima/notebook_utils.py

@@ -50,7 +50,7 @@ def psource(*functions):
         from pygments.lexers import PythonLexer
         from pygments import highlight
 
-        display(HTML(highlight(source_code, PythonLexer(), HtmlFormatter(full=True))))
+        display(HTML(highlight(source_code, PythonLexer(), HtmlFormatter(noclasses=True, style='monokai'))))
 
     except ImportError:
         print(source_code)